
Senior AI Engineer @google • Kauffman Fellow • ☸ • 🇵🇷 • ⛷ • 🏊🏾 • 🧘🏿♂️ • implacable • L∃∀N
San Francisco Bay Area
Joined May 2009
- Tweets12.6K
- Following1.1K
- Followers2.3K
- Likes12.9K
Marcos Polanco retweeted
I am in love with Gemini 4 Neon.
The model beats Opus 5.5 on KingBench 4 finishing just second to Fable 5.2.
Full review here: youtube.com/watch?v=eXeJNYoj…
Marcos Polanco retweeted
Gemini 4 Neon!! Fast. Cheap. Frontier.
Our 4th model in as many weeks and I have a feeling this will be a favorite for startups!!
Marcos Polanco retweeted
I am sorry to DeepMind for being harsh on them for the last 2 generations of models. Neon has redeemed them.
Gemini 4 Neon is currently the best coding model for everyday work. I prefer it over Opus 5.5, and I am probably going to daily drive this for a while. It has very interesting hidden capabilities.
If Antigravity can remove the 5-hour limits, that'd be great, as I run long-horizon tasks and that hinders my speed of shipping.
Using Carbon as reviewer and Neon as worker is getting me better results than Opus 5.5 + Fable 5.2.
Marcos Polanco retweeted
Matrices form nested worlds, each with its own laws of symmetry.
From raw arrays we factor into CR or the universal SVD:
A = Σ σᵢ uᵢ vᵢᵀ = UΣVᵀ.
Square matrices split into invertible and singular. Inside them sit the diagonalizable, then the normal matrices that commute with their transpose.
Symmetry tightens the circle: real eigenvalues, orthogonal diagonalization, projections that are their own square, and positive-definite matrices whose eigenvalues refuse to be negative. Orthogonal and permutation matrices sit at the center, eigenvalues pinned to the unit circle.
Every matrix, however wild, still possesses a clean singular-value story.
Marcos Polanco retweeted
More than 1.5 GW of chips are in China that shouldn’t be. Unwinding that web could reshape open-source AI.
“In our view, there’s more than one and a half gigawatts of chips in China that are not supposed to be there according to US law. How they get there, that whole web, that’s what’s going to unwind.”
“If NVIDIA can no longer support the open-source ecosystem, because that was Jensen’s primary goal, does that mean the open-source ecosystem falls away?”
“Now open source looks a lot weaker with NVIDIA down and out. Anthropic looks much more likely to get a complete monopoly or a duopoly. This bodes really well for Anthropic’s IPO next week.”
“They’re more diverse in what chips they use than any other company in the world.”
Marcos Polanco retweeted
It’s in our interest to take care of others. Self-centredness is opposed to basic human nature. In our own interest as human beings we need to pay attention to our inner values. Sometimes people think compassion is only of help to others, while we get no benefit. This is a mistake. When you concern yourself with others, you naturally develop a sense of self-confidence. To help others takes courage and inner strength.
Marcos Polanco retweeted
Love seeing two big advancements for the agentic era: massive context with TPU 8i Max and confidential inference with TPU 8tx!
And native PyTorch through TorchTPU is awesome too ❤️
Congrats Amin and team! :)
Thrilled to share a big milestone from our AI & Infrastructure team at @Google.
Today we’re expanding our 8th generation TPU portfolio with:
• TPU 8i Max: Built for massive context windows and the highest compute density per rack
• TPU 8tx: Purpose-built for secure, confidential inference
This marks the first time we’ve introduced four distinct TPU systems in a single generation, in one year.
For developers, hardware is only as good as the software stack. And we want you to be able to build with TPUs as easily and efficiently as we do at Google. You can now:
↳ Run native PyTorch & Hugging Face models via TorchTPU (zero code rewrites)
↳ Cut KV transfer latency >90% with open-source TPU Sync
↳ Eliminate idle compute waste with Agent Compute Engine
Read more about our newest TPUs on our blog: bit.ly/3VfETPR
I look forward to connecting with builders, researchers, engineers and organizations pushing the boundaries of AI and computing!
Marcos Polanco retweeted
Lean 4は、実際にはどのような仕組みで動いているのでしょうか?
その根底にある考え方は、驚くほどシンプルです。
「命題は型であり、証明はプログラム。そして、証明の検証は型チェックである。」
ある命題 P を証明したければ、型が P であるような項(term)を構築すればよいのです。定理証明器の仕事は、その項が本当に主張された型を持っているかを確認することです。
では、これは実際にはどのように実現されているのでしょうか?型チェッカーは、帰納法、等式、再帰、あるいは rw や simp のようなタクティクを、どのように扱っているのでしょうか?
今回、Rustで実装した最小限のLean 4風定理証明器 MiniLean4 と、その内部実装をゼロから解説する無料のインタラクティブな教材を公開しました。
learn.h5i.dev
この教材は、わずか5種類の項と8つの型付け規則から始まります。そこから少しずつ機能を追加し、実際に動作する証明チェッカーを構築していきます。
まず、型推論と型チェックを実装します。次に、簡約(reduction)と定義上の等価性(definitional equality)を導入します。これによって、カーネルは計算によって同じ結果になる式を等価なものとして認識できるようになります。
続いて、帰納型(inductive types)を実装します。 自然数から始めて、リスト、木構造、さらに一般的な帰納型の族(inductive families)へと拡張していきます。また、recursor(再帰子)を導出することで、再帰や数学的帰納法が内部的にどのように実現されているのかを理解します。
さらに、等式そのものも帰納型として実装し、少数の型付け規則と計算規則だけで、式の書き換え(rewriting)をどのように正当化できるのかを見ていきます。
最後に、タクティクエンジンを構築します。intro、apply、induction、rw といったタクティク自体を信頼する必要はありません。タクティクは証明項を構築するだけで、その正しさはカーネルが独立して検証します。
そのほかにも、宇宙レベル(universe levels)、代入、正規化、そして論理的な矛盾を避けるために型理論がどのように設計されているのかについても解説しています。
インタラクティブなデモも多数用意しました。実際のRust製カーネルをWebAssemblyにコンパイルしているため、ブラウザ上で直接証明を実行できます。
型付けの導出過程を確認したり、式が簡約される様子をステップごとに追ったり、タクティクによって証明状態がどのように変化するかを視覚的に確認したりできます。
このプロジェクトの目標は、定理証明を魔法のようなものではなく、普通のプログラミング言語の実装として理解できるようにすることです。
形式検証という言葉をよく聞くけど結局何をしているのか気になっている方、Lean 4を使ったことはあるけれど、by の裏側で何が起きているのか知りたい方、あるいは、型付け規則がなぜ形式検証の基盤になり得るのか興味がある方に、ぜひ読んでいただければと思います。
e-Bookはこちら!:learn.h5i.dev/
3 predictions:
1) In < 2 years most code will be written in Lean4 or equivalent
2) The final binary will be faster than Rust or C
3) Compiler architecture will change, with builtin intelligence and drastically more aggressive optimizations
btw if you can build a category leader open source project in ai engineering right now the market acquihire rate is ~$10-$100m per ai engineer.
you do not need to figure out a business model, you do not need GTM, you do not need funding.
just build things clankers want.
Models by @GoogleDeepMind scale better on cost than their rivals.
Gemini 4 Carbon is the prime example. As tasks get more demanding, every rival gets more expensive relative to Carbon.
For a typical task, GPT-6 Astra by @OpenAI costs 3.4× as much, Claude Opus 5.5 by @AnthropicAI costs 3.9× as much, and GPT-6.1 Sol is the cheaper option at about half of Carbon’s cost (0.54×).
Across the 5% most expensive sessions, Carbon’s position improves against all three. Opus edges up to 4.2×, Astra’s cost roughly doubles relative to Carbon, and Sol’s discount shrinks.
This in turn can translate into meaningful savings across real-world tasks such as building an Android security app, one of many long-horizon tasks that our Agent Arena users delegate to their agents.
As tasks get more complex, @GoogleDeepMind and @XiaomiMiMo are the biggest winners in the Agent Arena.
On cheaper tasks, @OpenAI and @AnthropicAI hold 10 of the 11 frontier spots. On the most expensive 5% of sessions, they hold just 7. The freed spots go to Google’s Gemini 4 Argon and Xiaomi’s MiMo-V2.6 Flash and Pro.
These shifts matter as sessions costing $100 or more are not uncommon across real-world use cases coming from Agent Arena. Even within the last 2 weeks, more than 1,000 sessions crossed that boundary, and choosing a suboptimal model might have had a meaningful impact on that final bill.
Marcos Polanco retweeted
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math.
It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training.
We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵
Blog: rikiyat.github.io/blog/weigh…
Readers added context they thought people might want to know
LLM was given weights explicitly as text in prompt and used code/forward passes (per authors' logs) to ablate Qwen looping direction, a known thinking-mode issue per model card. Primarily fixed repetition in that mode, not general 'natively' improving weights/math.
rikiyat.github.io/blog/weight-sp…
github.com/RikiyaT/lm_und…
huggingface.co/Qwen/Qwen3.5-0…
huggingface.co/RikiyaT/Qwen3.…
Marcos Polanco retweeted
Google’s Gemini 4 Carbon is the new #1 on the Artificial Analysis Intelligence Index v5, and second only to Claude Opus 5.5 on the Artificial Analysis Coding Agent Index
Gemini 4 Carbon scores 59 on Intelligence Index v5, driven by leading performance in complex reasoning, knowledge, and scientific tasks. It also has large gains in coding and knowledge work, particularly when compared to other Gemini models, but it is not leading in all these areas.
As with launch of Gemini 4 Argon, Gemini 4 Carbon is being rolled out to selected users and is not yet publicly available. Google is offering a 50% launch discount, with no confirmed end date. All our charts and figures use standard pricing for both Gemini 4 Carbon and Argon.
Key results for Gemini 4 Carbon with high reasoning:
➤ Gemini 4 Carbon leads the Intelligence Index v5, with the #1 score in 4 of the 9 Index evaluations: HLE-Diamond, Terminal-Bench-Science, AA-Omniscience, AutomationBench-AA. Carbon makes substantial gains elsewhere but is not leading in coding and knowledge work
➤ Behind only Claude Opus 5.5 on coding agents: With Antigravity CLI, Gemini 4 Carbon (high) scores 59 only 3 points behind Claude Code with Claude Opus 5.5 (max, 62) on the Artificial Analysis Coding Agent Index. Carbon ranks second on both Terminal-Bench 4.0 (64%) and FrontierCode v1.1 (51%), behind Claude Opus 5.5 (max) at 56%. Google has historically underperformed in this area and so this improvement stands out
➤ Improved agentic knowledge work, but not yet leading: On AA-Briefcase, Gemini 4 Carbon (high) scores 1609 Elo, a 126-point improvement over Gemini 4 Argon (high, 1483), but still behind models including Grok 4.7 (xhigh, 1638), Muse Wonder (max, 1659) and Kimi K3.1 Pro (1678). The largest gain comes from Presentation, up 293 Elo to 1598, alongside a 120-point improvement in Analytical Quality to 1671
➤ Lower Cost per Task than Claude Opus 5.5 for comparable overall intelligence: Gemini 4 Carbon costs $7.76 per Intelligence Index task, ~23% less than Claude Opus 5.5 (max, $10.04) but ~1.2x GPT-6 Astra (max, $6.25). It uses 123k output tokens per task, 29% fewer than Claude Opus 5.5 (max, 173k). At the same $4/$20 standard pricing, its Cost per Task is in line with Gemini 4 Argon ($7.68), while scoring 9 points higher on the Intelligence Index
➤ Leading AA-Omniscience v1.1 score, with low hallucination: Gemini 4 Carbon scores 54 on our recently updated AA-Omniscience v1.1, 6 points ahead of Claude Opus 5.5 (max, 48) and 8 points ahead of Gemini 4 Argon (46). Accuracy (% correct) rises 10 points over Argon to 58%, while its hallucination rate rises from 4% to 11%. This is still the second lowest of any model, behind only Argon, compared with 50% for both Claude Opus 5.5 (max) and GPT-6 Astra (max)
Key model details:
➤ Context Window: 1M tokens
➤ Multimodality: Text, image, video, and speech input, with text output
➤ Pricing: $4/$20 per 1M input/output tokens, the same as Gemini 4 Argon’s standard pricing, with a 50% discount. Cached input tokens receive a 95% discount ($0.20 per 1M)
Congratulations to the @GoogleDeepMind team!
Marcos Polanco retweeted
Gemini 4 Carbon is our next frontier model.
It’s state-of-the-art across cybersecurity and long-horizon capabilities.
Marcos Polanco retweeted
Meet Nex-N2.5 family, our latest open source agentic models.
Mini (35B) and Pro (397B) bring multimodal understanding and computer use. Max (1.6T) is a text-only MoE model built for complex reasoning, coding, and agent workflows.
Strong benchmark results in workflow automation and computer use:
⚡ Max scores 50.2 on AutomationBench v1.0.6—just 0.1 points behind Claude Opus 5
⚡ Pro scores 56.4 on OSWorld-2, ahead of Qwen3.8-Max at 46.7
With continuous action, visual feedback, and self-correction, NEX-N2.5 can work across Blender and CAD tools, handle expense reimbursements, and even play PC games—turning visual understanding into sustained, real-world execution.
🔗@huggingface
huggingface.co/nex-agi/Nex-N… huggingface.co/nex-agi/Nex-N… huggingface.co/nex-agi/Nex-N…
🔗@modelscope
modelscope.cn/models/nex-agi… modelscope.cn/models/nex-agi… modelscope.cn/models/nex-agi…
🔗 Github github.com/nex-agi/Nex-N2.5
🔗Website nex-agi.com/
Marcos Polanco retweeted
I'm happy to share that Mojo and MAX are now fully open - for development, extension, and commercial use! This includes the full Mojo programming language, but also the MAX graph compiler, MLRT runtime internals, and hardware abstraction logic.
Get your agents going!👇


























